| Summary: | cannot start netmount as net.eth0 could not start | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | codyzapp <codyzapp> |
| Component: | [OLD] baselayout | Assignee: | Gentoo's Team for Core System packages <base-system> |
| Status: | RESOLVED FIXED | ||
| Severity: | major | CC: | f5d8fd51ed1e804c9e8d0357e8614e0493b06e96, moonwick |
| Priority: | Normal | ||
| Version: | 2006.1 | ||
| Hardware: | x86 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
Copy of my /var/lib/init.d/deptree
Copy of my /var/lib/init.d/deptree (corrected) /var/lib/init.d/deptree broken |
||
|
Description
codyzapp
2006-09-23 02:42:58 UTC
I'm seeing basically the same problem. For me its ypbind that tells me it can't start due to net.eth0 failing to start. The cause for this is ypbind trying to start before net.eth0 even tried to start. It looks to me like its started later cause its running now - but still it yields those messages during boot. have you tried running: depscan.sh it should fix init.d dependencies. I had the exact problem, same error message, but I knew net.eth0 was started 'cause I had network & internet access. depscan.sh resolved my issue sorry for the delay... depscan.sh did not solve the issue for me, the messages still were shown after a reboot. Could you guys please attach your /etc/conf.d/net files? (In reply to comment #4) > Could you guys please attach your /etc/conf.d/net files? > # This blank configuration will automatically use DHCP for any net.* # scripts in /etc/init.d. To create a more complete configuration, # please review /etc/conf.d/net.example and save your configuration # in /etc/conf.d/net (this file :]!). config_eth0=( "dhcp" ) dhcp_eth0="nodns nontp nonis" During the boot process, do you see eth0 attempting to run dhcp and does it get and address successfully? I am having the same problem with a fresh install of Gentoo. It appears the dependencies for netmount have gone awry and it is being started in the default runlevel before net.eth0. I have RC_NET_STRICT_CHECKING set to NO in /etc/conf.d/rc so it should need to bring up eth0 before it considers the net service up and running (there is only one interface in this machine). It does start net.eth0 after it tries to start netmount, It runs dhcpcd and gets an IP address as expected, so when it has finished startup I do have access to my network, but my Samba shares are not mounted. When I noticed the problem I used the interactive startup mode to list the sequence of the scripts in the default runlevel. Here is the sequence on the problem machine: INIT: Entering runlevel 3 syslog-ng dbus hald netmount (ERROR: cannot start netmount as net.eth0 could not start) xdm net.eth0 numlock vixie-cron local Could people attach (not post as it's too big) their /var/lib/init.d/deptree files please. Also, if any have this problem, try running `depscan.sh -u` and see if that fixes. For whatever it's worth, I'm having what looks to be a very similar problem. In my case, I have several IP aliases listed in /etc/conf.d/net, and it appears that having more than two or three IPs listed causes net.eth0 to take long enough to run that it keeps other services which depend on it from starting. I'm not sure if it's noteworthy or not, but something interesting I'm noticing is that I never see net.eth0 mentioned during startup (it should list each of the IPs it's starting up) but I do see net.lo mentioned. Nonetheless, as soon as the machine is done booting, if I log in, ifconfig shows that all of the IPs were configured properly. I'll include my /var/lib/init.d/deptree in case it helps. For the record, I have not had the opportunity to run depscan.sh (as one reply suggests) because the machine is a server at a colo facility, and console access isn't convenient at the moment (for bringing up sshd by hand, since that's one of the services that inevitably fails to start.) Created attachment 100806 [details]
Copy of my /var/lib/init.d/deptree
Comment on attachment 100806 [details]
Copy of my /var/lib/init.d/deptree
(Uploaded file from wrong server. My bad.)
Created attachment 100810 [details]
Copy of my /var/lib/init.d/deptree (corrected)
(In reply to comment #6) > During the boot process, do you see eth0 attempting to run dhcp and does it get > and address successfully? > no it doesnt it just starts net lo nothing about net eth0 but i have internet so obviously it does... depscan doesnt fix the error either. It seems to have been a problem since the 2006.1 release. when i ran /etc/init.d/net.eth0 stop it stops eth0 sucessfully when i run start it starts it (hangs there never says OK like when it does when i run "stop" but if i close that terminal it kills internet.so if i dont close that terminal after running that it will be fine. (In reply to comment #4) > Could you guys please attach your /etc/conf.d/net files? > Just in case it helps, my /etc/conf.d/net file contains the usual comments at the top and the following three lines: config_eth0=( "dhcp" ) dhcp_eth0="release" dhcpcd_eth0="-t 10 -l 21600" (In reply to comment #8) > Could people attach (not post as it's too big) their /var/lib/init.d/deptree > files please. > > Also, if any have this problem, try running `depscan.sh -u` and see if that > fixes. > I made a copy of /var/lib/init.d/deptree then ran 'depscan.sh -u' as root. I tried to create a diff afterwards but there was no difference between the before and after files. I rebooted the computer and verified that the problem was not fixed. I had high hopes. Created attachment 100845 [details]
/var/lib/init.d/deptree
deptree from system with netmount error.
(In reply to comment #12) > Created an attachment (id=100810) [edit] > Copy of my /var/lib/init.d/deptree (corrected) You have the coldplug init script - try removing it as udev now handles coldplugging and this is the likely cause. (In reply to comment #17) > (In reply to comment #12) > > Created an attachment (id=100810) [edit] > > Copy of my /var/lib/init.d/deptree (corrected) > > You have the coldplug init script - try removing it as udev now handles > coldplugging and this is the likely cause. > how do i do this? i dont know where the coldplug init script is located.. (In reply to comment #9) > ... > I'm not sure if it's noteworthy or not, but something interesting I'm noticing > is that I never see net.eth0 mentioned during startup (it should list each of > the IPs it's starting up) but I do see net.lo mentioned. Nonetheless, as soon > as the machine is done booting, if I log in, ifconfig shows that all of the IPs > were configured properly. > ... This also applies to comment #13 If you are running a desktop environment you may not see net.eth0 start up in the console because the xdm script is started first. On my new install that cannot start netmount I see the order it runs init scripts is somewhat different than it is for an older install I have on a different machine. As I have listed in comment #7, the machine with the problem tries netmount, then runs the xdm script which starts gdm and switches me to the X display, then it runs the net.eth0 script. For me to see the output "Starting net.eth0" "Starting dhcpcd" "obtaining IP address" etc. I have to press CTRL+ALT+F1 to switch back to the first text console after it switches me to the X display. You can switch back to X with CTRL+ALT+F7 depscan.sh -u didn't solve the problem. But changing RC_PARALLEL_STARTUP="no" to "yes" in /etc/conf.d/rc solved it. Using parallel boot, net.eth0 is start first before netmount. Created attachment 103163 [details]
broken
broken deptree
This happens on 2 machines that are fresh installs using sys-apps/baselayout-1.12.6 This should be fixed in baselayout-1.13.0_alpha11. Re-open if you disagree. |
